|Publication number||US5095512 A|
|Application number||US 07/261,956|
|Publication date||Mar 10, 1992|
|Filing date||Oct 24, 1988|
|Priority date||Oct 24, 1988|
|Also published as||CA2001405A1, CA2001405C, DE68928046D1, DE68928046T2, EP0441896A1, EP0441896A4, EP0441896B1, WO1990004838A1|
|Publication number||07261956, 261956, US 5095512 A, US 5095512A, US-A-5095512, US5095512 A, US5095512A|
|Inventors||Lawrence G. Roberts, Dennis L. Hunt|
|Original Assignee||Netexpress Systems, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (6), Non-Patent Citations (6), Referenced by (17), Classifications (8), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to image data processing and more particularly to converting between image data of a plurality of differing formats. Specifically, the invention relates to the conversion between defined forms of standardized image data, wherein one form involves single-line coding and another form involves multiple-line coding in a manner optimized for speed of processing. The invention has particular application in facsimile data transmission and in bit-mapped image reproduction, such as in connection with electrostatic (laser) graphics printers.
Image data is an important data form for conveying information of both text and graphic content. The current market for laser printers alone is valued at more than one billion dollars. The facsimile market has a similar large market value.
Image data is typically coded as either a compressed code or a bit-mapped code. A single standard page-sized bit-mapped data image having a resolution of 400 lines per inch requires sixteen million bits of storage. Transmission of a bit-mapped image a bit at a time over conventional telephone-grade 3000 Hz bandwidth communication media is considered to be prohibitively time-consuming.
In order to reduce the size of image data, various compression techniques have been adopted. In the facsimile art, for example, image data is reduced to one of several types of compressed code form, such as Group III (Modified Huffman or MH) or Group IV (MMR) prior to transmission.
Frequently it is necessary to convert compressed image data from one compressed code form to another compressed code form in order to overcome the lack of compatibility among types of equipment. In the past it has been considered that conversion between Group III and Group IV coded images is so complex that it could not be done except by creation of a complete bit-mapped image version of the source data and applying conventional facsimile encoding process to obtain the image in the destination code format. Nevertheless, a straight-forward manipulation of a full-page bit-mapped image at 400 lines per inch requires processing of 16 millions bits of data. Such a process is extremely time-consuming and cumbersome. A bit-mapped data image is typically not further manipulated, except that it may be used to reproduce an image output on an electrostatic printer or the like. Thus, conventional facsimile and printing technology has encountered a barrier in the trade-off between image resolution and the speed of image processing. Whereas standard facsimile image data is considered too complex for any sort of meaningful image data manipulation, bit-mapped image data is considered to be simply too massive to be manipulated efficiently. Nevertheless, standardized facsimile image data is attractive because facsimile is becoming universally acceptable as a mode for transferring information. Thus, needs exist to provide better techniques for converting facsimile coded compressed image data among various formats, to speed the process of image reproduction and to process the image information in general. Further unexpected benefits might also accrue with the solution to these problems.
According to the invention, bit-mapped image data, as well as selected forms of compressed image data, are converted between one another by converting data through an intermediate data and code structure wherein each image is represented by a list of sublists, each sublist being itself a list of numerical values representing run lengths between transitions from one type of picture element (pel) to a second and opposite type of picture element (pel) (e.g., a black pel and a white pel). Hence the intermediate code data structure according to the invention is termed a transition list or TL code or data structure. The data structure and coding rules are defined by rules relating to code type, code word length and code word interdependence. A specific conversion process specifies that the transition list be composed of sublists wherein each represents a scan line. In a particular embodiment, the rules specify 1) that the intermediate code be a run length-type code of each scan line of the image, 2) that the code words be of fixed length and 3) that the image be stored in a (two-dimensional) serial array wherein each scan line is of a length dependent only on the number of transitions of run lengths. The preferred form of intermediate code is a fixed word-length transition list, that is, a serial list of values representing run lengths of either "black" picture elements or "white" picture elements stored serially in digital data format having a fixed word length, wherein the last value is repeated three times to indicate end of line. Alternatively, a scan line header word may specify scan line length.
A specific conversion process according to the invention specifies that the transition list be composed of sublists representing a scan line, each scan line consisting of a header word indicating line length from zero to a value indicating page width, and a grouping (in pairs) of words specifying run lengths of white pels and run lengths of black pels. A rule specifies 1) that a full transition list (a page) have a scan line header word specifying the number of transitions (an even number), 2) that the first run length code word be the incremental (rather than cumulative) number of consecutive white picture elements (stored in a two-byte word), 3) that the second run length code word be the incremental (rather than cumulative) number of consecutive black picture elements (stored in a two-byte word), 4) that the last run length code word be a black run length code, thus pairing white and black run length units and skipping the final white code word in each scan line, 5) that the end-of-segment (or end-of-image) marker be a negative one (-1) stored in a two-byte word, 6) that the segment header word be a two-byte word specifying the address of the next segment of memory, or that there is no further segment, and finally 7) that the image header consist of five two-byte words specifying a) height of the image in lines, b) width in maximum number of picture elements, c) vertical origin in terms of number of picture elements up to origin from the bottom left image corner, d) horizontal origin in terms of number of picture elements right to the origin from the bottom left image corner, and e) proportional width in terms of number of picture elements right of the origin to the next character (which is used to control overlapping of images). In segmented images, pointers may be provided together with a trailing end-of-segment marker, an optional segment header word pointing to the next segment or end of image, and an image header specifying height, width, location and proportion of the image.
The invention further contemplates processes for converting from MH to TL, from TL to MH, from MMR to TL, from TL to MMR, from text to TL, from TL to printable bit-mapped image, from TL to displayable bit-mapped image, and from scanned bit-mapped image to TL.
Significantly, the intermediate code according to the invention is of a form sufficiently compressed and sufficiently modularized that important and desired intermediate processes can be applied to it, such as image merging, image scaling and image cut and paste. Hence, the processes according to the invention contemplate TL merge, TL scaling and TL cutout.
In the course of developing the present invention it was noted that a typical compressed data image of a page of text contains a limited number of transitions between "black" and "white" in the course of a serial scan of a pictorial image. For example, a typical image with a resolution of 300 to 400 lines per inch contains only about 160,000 transitions. Recognizing these and other characteristics makes it possible to produce a homogeneous image output suitable for either facsimile transmission or for hard-copy image reproduction. It could well emerge that a standard form of information may emerge comprising a hybrid of electronic data and electronic image. This invention permits rapid processing and merging of such a form into a final hard copy form, and it permits transmission of an electronic image of such information in a standard facsimile format.
The invention has been found to be capable of processing a page of information at a rate of 4 seconds per page using conventional facsimile processing hardware such as the Am7970A compression expansion processor chip built by Advanced Micro Devices of Sunnyvale, California in conjunction with an 80286 microprocessor chip of Intel Corporation of Santa Clara, California. A conventional process using the same hardware would require about 60 seconds per page. Conventional image processing employing faster hardware, such that based on as a 68020 microprocessor chip of Motorola Corporation of Chicago, Illinois, are able to achieve a speed of about 15 seconds per page. Nevertheless, the present invention is able to produce its results with less than one one-hundredth of the processing power which would be required for full bit-mapped image processing.
FIG. 1 is a block diagram of a specific embodiment of the invention illustrating data conversion elements, data storage elements, data manipulation elements and data transmission and reception elements.
FIG. 1A is a flow chart for converting a first image to a second image by way of the apparatus of FIG. 1.
FIG. 2 is a simplified block diagram of a digital data processing apparatus for use in an apparatus according to the invention.
FIG. 3 is a diagram illustrating a specific embodiment of a transition list as stored in a block of a digital memory.
FIGS. 4A and 4B together are a flow chart of a process for converting transition list data into MMR data.
FIG. 5 is a flow chart of a process for converting transition list data into MH data.
FIG. 6 is a flow chart of a process for converting transition list data into MR data.
FIGS. 7A, 7B and 7C together are a flow chart of a process for converting MMR data into transition list data.
Referring to FIG. 1 there is shown a block diagram of a specific embodiment of an encoding/decoding apparatus 10 in accordance with the invention. The encoding/decoding apparatus 10 comprises an encoder 12, a modem (and transmission medium) 14 and a decoder 16. The modem 14 as coupled to a transmission medium is illustrated for the sake of completeness and is not an essential element to the understanding of the invention.
The modem (and transmission medium) 14 comprise for example an MMR modulator 18 coupled to a suitable transmission link 20, such as a 64 KBS transmission link suited to carrying standard MMR modulation as defined in CCITT Recommendation T.6 "FACSIMILE CODING SCHEMES AND CODING CONTROL FUNCTIONS FOR GROUP 4 FACSIMILE APPARATUS," Fascicle VII.3 - Rec. T.6 (Malaga-Torremolinos, 1984). The transmission link 20 is coupled to an MMR demodulator 22. At the transmission end the MMR modulator 18 is coupled to receive MMR data from a suitable data buffer or memory, such as a two-port MMR memory 24. At the receiving end the MMR demodulator 22 is coupled to a suitable two-port MMR memory 26 which is operative to capture received and demodulated MMR data in order to buffer it for further processing in accordance with the invention.
It is contemplated that each terminal of a facsimile network employing an apparatus in accordance with the invention comprise the three elements of an encoder 12, a modem 14 and a decoder 16. However it is to be understood that such an apparatus is capable of communication with any other Group 4 apparatus through its modulator 18 and demodulator 22.
In accordance with the invention the encoder 12 is suited to receive at its input 28 digital information which does not have a one-for-one correspondence with a target code such as MMR (CCITT Group 4 facsimile code) and to analyze and convert that source information through a transition list processor 30 into a transition list code, and thereafter to process the information in the transition list code through a transition list processor 32 and/or convert the transition list code through a transition list to output code converter 34 into a target code such as MMR code The input digital information of an encoder 12 in accordance with the invention may be MH code (CCITT Group 3 facsimile code as defined CCITT Recommendation T.4 "STANDARDIZATION OF GROUP 3 FACSIMILE APPARATUS FOR DOCUMENT TRANSMISSION," Fascicle VII.3 - Rec. T.4, Geneva, 1980, amended at Malaga-Torremolinos, 1984), bit-mapped image data, text code, such as USASCII or EBCDIC, or eventually even an image description in a printer description language code such as DDL (Data Description Language of Imagen Corporation) or PostScript of Adobe Systems.
To this end the input 28 is coupled to an input memory 36 which in turn is coupled to the input to transition list processor 30, the output of which is coupled to a first transition list memory 38. The first transition list memory 38 is coupled to the transition list code to transition list code processor 32, which in turn is coupled to the second transition list memory 39. The second transition list memory is coupled to the transition list to output code (MMR) converter 34, which in turn is coupled to provide output code to the MMR memory 24 of the modem 14. First control means 40 are provided for controlling the nature of processes carried out by the transition list. processor 32, such as merge, scaling or cutout. The various memory means are provided for temporary storage of code during processing. It is contemplated that the apparatus 12 will operate in near real time, and that the memory means are adapted to high throughput applications. It is further contemplated that the memory means may be embodied in a signal physical memory unit, such as a semiconductor random access memory device, or into a plurality of memory units serving as two-port input-output buffer memories. It is still further contemplated that the processors 30, 32, and 34 may be embodied in a programmable microprocessor unit operative to execute computer programs for performing the input to transition list conversions, the transition list to transition list manipulations and the transition list to output code conversions. Suitable microprocessors for commercial applications of the invention are the Motorola 68000 series microprocessors. The operations of the microprocessor may be augmented by conventional facsimile processing hardware such as the Am7970A compression/expansion processor chip built by Advanced Micro Devices of Sunnyvale, California in conjunction with an 80286 microprocessor chip of Intel Corporation of Santa Clara, California.
Further in accordance with the invention the decoder 16 is suited to convert received source code from the modem 14, typically in a code such as MMR, through the mechanism of a source code to transition list converter 42 into a transition list code and thereafter to process the transition list code in a transition list code processor 44 under control of a second control apparatus 45 and/or convert the transition list encoded data via a transition list code to output code processor 46 into a target code for transmission to an output apparatus 48, such as a printer, a display or another facsimile transmitter. The target code may be a bit-mapped image, text, MH code or a printer description language code. Images for display may be bit-mapped code or text code. Images for printing may be PDL code, text code or bit mapped code. Images for transmission into another media such as according to a different facsimile standard may be converted into a facsimile code such as MH code or some other compressed image code. It is important to recognize that according to the present invention, there need not be a direct correlation between the source code and the target code, so long as a conversion exists whereby image data can be converted into and out of transition lists
In order to provide temporary storage conversion and manipulation processes, a third transition list memory 50, a fourth transition list memory 52 and an output memory 54 are provided between the processors as above. It is contemplated that many of these functions are shared by the same physical elements and further that subsets of the apparatus as shown in FIG. 1 may be provided for dedicated and special purpose functions where not all of the options as illustrated are needed or used.
FIG. 1A illustrates a summary of the steps implemented by the apparatus 10 of FIG. 1 used to convert a first image described by a first graphics image code to a corresponding second image described by a second graphics image code. First, step A' analyzes the image represented in the first graphics image code. This analysis step identifies a first scan line representation for each scan line of the first graphics image code. This analysis step may involve analysis of an entire image to identify a first scan line representation. The next step, step B', converts the first scan line representation into a line of code of a third code sufficiently articulate to capture the entire denotation of the first scan line representation. Thereafter, in step C', the process accumulates a sufficient number of line codes of the third code to construct a translation unit of the second graphics image code. This may involve construction of a block code from the translation code. Thereafter, the translation code may be manipulated optionally (Step D'). Rotations, translations, etc. are permitted on the basis of the translation code. Finally, the process is completed with step E' when the second image, described by the second graphics image code, is constructed from the translation unit. Significantly, no one-to-one code word conversion correspondence is necessary to convert the first image to the second image. Moreover, there is allowed the conversion of one type of bit-mapped image to another type of bit-mapped image and the manipulation (e.g., rotation, translation, etc.) of the structure of the image through processing of the intermediate or translation code.
FIG. 2 is a block diagram of a typical microprocessor-based apparatus 10 incorporating the features of the present invention. Other embodiments are within the skill of the ordinary artisan in this art. The apparatus 10 comprises a data and control bus 56 to which are coupled a central processing unit, i.e., a microprocessor 58, input/output channels 60, a data input interface 62, a data output interface 64, mass storage 66, random access memory 68, read only memory 70 and a special processor 72. A data source 74 is coupled to provide source data through the input interface 62 to the random access memory 68 under control of the microprocessor 58, and a data destination 76 is coupled to receive destination data via the output interface 64 under control of the microprocessor 58. The programs for controlling data translation and manipulation, as well as operating system functions, are stored in the read only memory 70. A control device 78, such as a terminal or remote data link, provides overall command and control via the input/output interface 60. Special processing of facsimile data is, for example, handled by the dedicated processor 72 coupled in accordance with the specifications for the dedicated processor to the bus 56 and the microprocessor 58. It should be understood that other structures may be employed, such as a structure employing pipeline memory, cache memory or other structures determined to be suited to processing of facsimile image data.
Referring to FIG. 3, there is shown a sample of the data structure of a transition list 80 in accordance with the invention. For convenience, the data structure is shown as consecutive addressable locations in a memory space, each location comprising 16 bits of storage. In accordance with a specific embodiment of the invention, a transition list 80 comprises a superheader 82, a segment header 84, and a plurality of segments or transition sublists 86, 88, 90, 92, 94, 96, and 98, and an end of segment block marker 100. The transition list 80 represents the complete description of an image independent of the compressed code or bit image representation of the image. Each of the segments or sublists 86, 88, 90, 92, 94, 96, and 98 is a fixed bit-length number representing the number of consecutive bits in the scan line of an image without transition, the absence of an image registration, and the presence of an image registration, e.g., the number of consecutive bit transitions from white to black and from black to white in a black and white image. (Color images may be represented in a similar manner using for example conventional three or four-pass representations of a single image frame.)
An alternative transition list format is illustrated by the example shown in the following table. The format differs in that the last code is tripled to indicate an end to the list. In the data field, a 0 represents a white, a 1 represents a black.
______________________________________Element Start Address______________________________________1 02 103 214 225 356 997 17288 17289 1728______________________________________
According to this representation of the transition list, it is unnecessary to know the number of elements, i.e., the length of the transition list, as the transition list is provided with a terminating element, a repetition of the last transition three times. Each element is a 16-bit word, which allows representations of addresses from 0 to 65,536.
FIG. 4 is a flow chart of a representative conversion from a transition list (TL) data to MMR code, the standard for CCITT Group IV facsimile. Herein it is assumed that a transition list has been created by any process. The formulas in the flow chart employ the syntax of the C programming language. The reference designations are the same as those used in the CCITT specification for positions: a0, a1, b1, and b2.
Starting with initialization of the color (=white), a0 to 0, a1 to the current value plus 1 and b1 to the reference value plus 1 (Step A), the test is made for the "pass" mode (Step B). If it is the pass mode, then the pass code is inserted (Step C) and a0 is set to b2, ref+ is set to 2 and b1 is set to the value ref (Step D). The process then proceeds to the test for end of line (Step E). If a0 is not less than the line length the process begins again at Step B; otherwise it terminates (F).
If the pass mode test (Step B) yields a negative, then the Vertical test is applied (Step G). If affirmative, the proper vertical code is inserted and a0 is set to a1 (Step H). Vertical position is tested from position 0 (Step I) to positions 1-3 (Step J). If at position V0, the ref value is incremented (Step K) and Step E is repeated. If at positions Vl through V3, several steps are invoked. The value ref is incremented (Step L), then tested against a1 for position (Step M), which if less, sets ref+1 to 2 (Step N) and invokes the color switch (Step O); otherwise the color switching step is invoked immediately. If the vertical right tests are negative, it is presumed that the position is vertical left 1 through 3. Thereupon, ref-1 is tested against a1 (Step P), which if greater decrements ref (Step Q) or if lesser, it increments ref (Step R) and then proceeds to Step 0, the color switching step. The color switching step leads to Step E.
If the vertical test is negative (Step G), the horizontal processing steps are invoked. Into the bit stream are inserted the horizontal code, the run length codes for length a1 - a0 at the current color and the run length codes for length ((current+1) - a1) for the opposite color (Step S). Thereafter, a0 is incremented by current plus 1, current plus 1 is set to 2 and a1 is set to current (Step T). While b1 remains less than a0, ref plus 1 is set to 2 and b1 is set to ref (Step U). Thereafter Step E is invoked. The process is repeated until the entire transition list is translated to MMR code. The last element of a transition list is repeated three times to signal the end of a scan line. At the completion of all transition lists for all scan lines is a block or segment, an end of block (EOB) code is inserted in the bit stream (Step V) in accordance with established MMR standards.
FIG. 5 is a flow chart for illustrating the conversion of a transition list to MH (Modified Huffman) code in its compressed version. Initially, an End of Line (EOL) code is inserted in the bit stream (Step W) after which the current relative position ("position") is set to zero and the color is set to white (Step X). A series of iterative steps begin. The value "len" is set to the value of (current+1) less the value of "position" and the value of "position" is set to "current" (Step Y). Thereafter the run length code or codes of length "len" and color value stored as "color" are inserted in the bit stream (Step Z). (The two possible values of "color" are black and white.) The colors are then switched (Step AA) and the value of "position" is tested against "line length" (Step AB) to determine if the line is completed. If not, steps Y through AB are repeated until the line is completed. When the end of a block is reached, preferably six End of Line (EOL) codes are inserted into the bit stream as a trailer (Step AC).
MMR is essentially a two dimensional code whereas MH is a one dimensional code. A third code exists, called MR, which is a bit mapped coding scheme which can be either a one or two dimensional code. FIG. 6 is a flow chart useful for understanding how to convert TL to MR. The first step is to test whether the TL is a one-dimensional or two dimensional code (Step AD). This is typically information available in the header of the TL code (Step AD1). If the code is one dimensional, then it is encoded using the procedures established for MH, above (FIG. 5) (Step AE). If the code is two dimensional, then it is encoded using the procedures established for MMR, above (FIG. 4) Step AF). At the conclusion of either of these procedures, at the end of a block, preferably six End of Line (EOL) codes are inserted into the bit stream as a trailer (Step AG).
FIG. 7 shows the process for converting MMR to TL, also called decoding MMR. It may be compared with the process described in connection with FIG. 4, relating to encoding MMR. Using the previous conventions for symbols, The process is started (Step AH) and the various initial values and pointers are initialized, with the initial color being set to WHITE (Step AI). The process then enters a loop. The next step is to find, from the input data stream, the next two dimensional code which satisfies one of several criteria: VL3, VL2, VL1, V0, VR1, VR2, VRE3, HOR, PASS, EXTENSION or EOB (Step AJ). The codes are then tested. If the code is V0 (Step AK), the a0 value is initialized to a first value, as noted in the flow chart, the b1 pointer is incremented, and the colors are switched (Step AL) completing the coding phase. If the code is VLI through VL3 (Step AM), the a0 value is initialized to a second value, as noted in the flow chart and the colors are switched (Step AN). Thereafter the b1 pointer minus 1 is tested against the initial value a0 (Step AO) and the b1 pointer is either decremented (Step AP), or incremented (Step AQ) toward a value related to a0, as noted in the flow chart. This also completes the coding phase.
If the code is VR1 through VR3 (Step AR), the a0 value is initialized to a third value, as noted in the flow chart and the colors are switched (Step AS). Thereafter the b1 pointer is tested against the initial value a0 (Step AT) and the b1 pointer is either incremented by 2 (Step AU) or considered complete. This also completes the coding phase.
If the code is PASS (Step AV), the a0 value is initialized to a fourth value, the bl pointer is incremented by 2 (Step AW), also completing the coding phase.
If the code is HOR for horizontal (Step AX), certain codes related to the horizontal coding are invoked. Optionally, the horizontal make-up codes are found, and then the terminating code for the current color is found. Thereafter, the a0 value is set as noted in FIG. 7 for both colors during a loop under which certain conditions are true (Step AY). This completes the horizontal portion of the coding phase.
If the code is an EOB (End of Block) code (Step AZ), the end of the sequence is noted (Step BA). This also completes the coding phase. If the code is an extension code (EXT) (Step BB), then the encoding executed is related to an optional uncompressed mode (Step BC). Absent any other valid code, the result is an error signal (Step BD).
At the end of every coding phase, the value ao is tested against the total line length (Step BE). If it is less, then the process is repeated from Step AJ. If it matches line length, then the line is done and coding is complete. The process returns the transition line suited to conversions to other coding schemes.
The Tables attached hereto provide details on one embodiment of the invention. Table A illustrates and outlines a procedure for implementing TL to MMR conversion. Table B provides detailed source code listing of the decoding and encoding processes in accordance with one embodiment of the invention.
The invention has now been explained with reference to specific embodiments. Other embodiments will be apparent to those of ordinary skill in this art. It is therefore not intended that this invention be limited, except as indicated by the appended claims. ##SPC1## ##SPC2## ##SPC3## ##SPC4## ##SPC5## ##SPC6## ##SPC7## ##SPC8## ##SPC9## ##SPC10## ##SPC11## ##SPC12##
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4602383 *||May 1, 1984||Jul 22, 1986||Nippon Telegraph & Telephone Public Corporation||Image data compression system|
|US4610027 *||Dec 30, 1983||Sep 2, 1986||International Business Machines Corporation||Method for converting a bit map of an image to a run length or run end representation|
|US4646356 *||Jun 29, 1984||Feb 24, 1987||International Business Machines Corporation||Method for converting a bit map of an image to a run length or run end representation|
|US4783834 *||Feb 20, 1987||Nov 8, 1988||International Business Machines Corporation||System for creating transposed image data from a run end or run length representation of an image|
|US4858017 *||Jan 22, 1988||Aug 15, 1989||The Trustees Of Columbia University In The City Of New York||System and method for hierarchal image encoding and decoding|
|US4876607 *||Mar 31, 1982||Oct 24, 1989||International Business Machines Corporation||Complex character generator utilizing byte scanning|
|1||CCITT Recommedation T.4, "Standardization of Group 3 Facsimile Apparatus for Document Transmission", Fascicle VII.3, Geneva, 1980, amended at Malaga-Torremolinos, 1984.|
|2||*||CCITT Recommedation T.4, Standardization of Group 3 Facsimile Apparatus for Document Transmission , Fascicle VII.3, Geneva, 1980, amended at Malaga Torremolinos, 1984.|
|3||CCITT Recommendation T.6, "Facsimile Coding Schemes and Coding Control Functions for Group 4 Facsimile Apparatus", Fascicle VII.3, Rec. T.6, (Malaga-Torremolinos, 1984).|
|4||*||CCITT Recommendation T.6, Facsimile Coding Schemes and Coding Control Functions for Group 4 Facsimile Apparatus , Fascicle VII.3, Rec. T.6, (Malaga Torremolinos, 1984).|
|5||*||Roy Hunter et al., International Digital Facsimile Coding Standards, 7 1980, 854 867.|
|6||Roy Hunter et al., International Digital Facsimile Coding Standards, 7-1980, 854-867.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US5276526 *||Aug 26, 1991||Jan 4, 1994||Canon Kabushiki Kaisha||Decoding device for two-dimensional codes by using first or second reference line signals|
|US5611067 *||Mar 29, 1993||Mar 11, 1997||Kabushiki Kaisha Toshiba||Nonvolatile semiconductor memory device having means for selective transfer of memory block contents and for chaining together unused memory blocks|
|US5774119 *||Sep 18, 1997||Jun 30, 1998||International Business Machines Corporation||Graphical interface method, apparatus and application for selection of target object|
|US5774120 *||Aug 14, 1996||Jun 30, 1998||International Business Machines Corporation||Refresh and select-all actions in graphical user interface|
|US5774634 *||Sep 22, 1994||Jun 30, 1998||Canon Kabushiki Kaisha||Image processing method and apparatus|
|US5781193 *||Aug 14, 1996||Jul 14, 1998||International Business Machines Corporation||Graphical interface method, apparatus and application for creating multiple value list from superset list|
|US5784057 *||Aug 14, 1996||Jul 21, 1998||International Business Machines Corporation||Dynamically modifying a graphical user interface window title|
|US5801775 *||Jul 17, 1996||Sep 1, 1998||Nec Corporation||Moving picture compression using cache memory for storing coding instructions|
|US5818444 *||Aug 14, 1996||Oct 6, 1998||International Business Machines Corporation||Method, apparatus and application for object selective but global attribute modification|
|US5867157 *||Aug 14, 1996||Feb 2, 1999||International Business Machines Corporation||Graphical interface method, apparatus and application for creating and modifying a list of values with multiple components|
|US5872568 *||Aug 14, 1996||Feb 16, 1999||International Business Machines Corporation||Application and method for creating a list from pre-defined and user values|
|US5890188 *||Oct 9, 1997||Mar 30, 1999||Kabushiki Kaisha Toshiba||Nonvolatile semiconductor memory device having means for selective transfer of memory block contents and for chaining together unused memory blocks|
|US5917947 *||Jul 11, 1994||Jun 29, 1999||Canon Kabushiki Kaisha||Image processing method and apparatus permitting use of PDL in compression memory|
|US6104395 *||Aug 14, 1996||Aug 15, 2000||International Business Machines Corporation||Graphical interface method, apparatus and application for opening window of all designated container objects|
|US6195096||Aug 14, 1996||Feb 27, 2001||International Business Machines Corporation||Graphical interface method, apparatus and application for creating and modifying a multiple-value text list|
|US6600840||Jul 21, 1997||Jul 29, 2003||International Business Machines Corporation||Image format conversion such as photometric, rotation, cropping, padding, scaling, dithering, bit padding, grayscale and color transformation, encoding and decoding using a plurality of filters|
|EP0801362A2 *||Apr 1, 1997||Oct 15, 1997||OcÚ-Technologies B.V.||Method and apparatus for converting image representation format|
|U.S. Classification||382/245, 358/426.13|
|International Classification||G06T9/00, H04N1/413|
|Cooperative Classification||G06T9/005, H04N1/413|
|European Classification||G06T9/00S, H04N1/413|
|Oct 24, 1988||AS||Assignment|
Owner name: NETEXPRESS SYSTEMS, INC, (A NEVADA CORPORATION), C
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ROBERTS, LAWRENCE G.;HUNT, DENNIS L.;REEL/FRAME:004962/0837
Effective date: 19881020
|Nov 3, 1993||AS||Assignment|
Owner name: SILICON VALLEY BANK, CALIFORNIA
Free format text: SECURITY INTEREST;ASSIGNOR:NET EXPRESS, INC.;REEL/FRAME:006756/0200
Effective date: 19931008
|Feb 2, 1995||AS||Assignment|
Owner name: NET EXPRESS, INC., VIRGINIA
Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:SILICON VALLEY BANK;REEL/FRAME:007322/0481
Effective date: 19931008
|Sep 11, 1995||FPAY||Fee payment|
Year of fee payment: 4
|Jul 30, 1999||FPAY||Fee payment|
Year of fee payment: 8
|Aug 19, 2003||FPAY||Fee payment|
Year of fee payment: 12